# Material Count¶

This module supports the virtual dimensionality measure.

See the file test_vd.py for an example.

## HSI Image Vitual Dimensionality Measure (HfcVd)¶

### Function¶

pysptools.material_count.vd.HfcVd(M, far='default')[source]

Computes the vitual dimensionality (VD) measure for an HSI image for specified false alarm rates. When no false alarm rate(s) is specificied, the following vector is used: 1e-3, 1e-4, 1e-5. This metric is used to estimate the number of materials in an HSI scene.

Parameters: M – numpy array HSI data as a 2D matrix (N x p). far – list [default default] False alarm rate(s).
Returns: python list
VD measure, number of materials estimate.

References

C.-I. Chang and Q. Du, “Estimation of number of spectrally distinct signal sources in hyperspectral imagery,” IEEE Transactions on Geoscience and Remote Sensing, vol. 43, no. 3, mar 2004.

J. Wang and C.-I. Chang, “Applications of independent component analysis in endmember extraction and abundance quantification for hyperspectral imagery,” IEEE Transactions on Geoscience and Remote Sensing, vol. 44, no. 9, pp. 2601-1616, sep 2006.

### Class¶

class pysptools.material_count.HfcVd[source]

Computes the vitual dimensionality (VD) measure for an HSI image for specified false alarm rates.

count(M, far='default', noise_whitening=False)

Computes the vitual dimensionality (VD) measure for an HSI image for specified false alarm rates. When no false alarm rate(s) is specificied, the following vector is used: 1e-3, 1e-4, 1e-5. This metric is used to estimate the number of materials in an HSI scene.

Parameters: M – numpy array HSI data as a 2D matrix (N x p). far – list [default default] False alarm rate(s). noise_whitening – boolean [default False] If True noise whitening is applied before calling HfcVd, doing a NWHFC.
Returns: python list
VD measure, number of materials estimate.

References

C.-I. Chang and Q. Du, “Estimation of number of spectrally distinct signal sources in hyperspectral imagery,” IEEE Transactions on Geoscience and Remote Sensing, vol. 43, no. 3, mar 2004.

J. Wang and C.-I. Chang, “Applications of independent component analysis in endmember extraction and abundance quantification for hyperspectral imagery,” IEEE Transactions on Geoscience and Remote Sensing, vol. 44, no. 9, pp. 2601-1616, sep 2006.

## Hyperspectral Signal Subspace Identification by Minimum Error (HySime)¶

### Function¶

pysptools.material_count.vd.hysime(y, n, Rn)[source]

Hyperspectral signal subspace estimation

Parameters: y – numpy array hyperspectral data set (each row is a pixel) with ((m*n) x p), where p is the number of bands and (m*n) the number of pixels. n – numpy array ((m*n) x p) matrix with the noise in each pixel. Rn – numpy array noise correlation matrix (p x p)
Returns: tuple integer, numpy array
• kf signal subspace dimension
• Ek matrix which columns are the eigenvectors that span the signal subspace.
Jose Nascimento (zen@isel.pt) & Jose Bioucas-Dias (bioucas@lx.it.pt) For any comments contact the authors

### Class¶

class pysptools.material_count.HySime[source]

Hyperspectral signal subspace identification by minimum error.

count(M)

Hyperspectral signal subspace estimation.

Parameters: M – numpy array Hyperspectral data set (each row is a pixel) with ((m*n) x p), where p is the number of bands and (m*n) the number of pixels.
Returns: tuple integer, numpy array
• kf signal subspace dimension
• Ek matrix which columns are the eigenvectors that span the signal subspace.
Reference:
Bioucas-Dias, Jose M., Nascimento, Jose M. P., ‘Hyperspectral Subspace Identification’, IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 46, NO. 8, AUGUST 2008.